Home

Welcome to the HEIG wiki. HEIG is a statistical framework for efficient joint analysis of imaging and genetic data. It is able to conduct voxelwise genome-wide association analysis using unrelated subjects for images and produce a minimal dataset of summary statistics to share with the community for secondary analyses. HEIG provides an estimator for voxelwise heritability, genetic correlations between pairs of voxels, and cross-trait genetic correlations between voxels and non-imaging phenotypes, only using the mininal dataset of summary statistics and a population-matched LD matrix and its inverse. More analysis modules will be supported in the near future.

Currently, HEIG has six major analysis modules, including kernel smoothing, functional PCA, summary statistics preprocessing, voxelwise GWAS, LD matrix construction, and heritability and (cross-trait) genetic correlation analysis. We also provide an introduction to visualizing the analysis results for the human brain. The FAQ answers common questions on HEIG.

To replicate the analyses showed in the tutorial, you can download the example data at Zenodo. The total file size is 1.8 GB.